Is it acceptable to extend results from an arXiv paper that has not appeared in a journal?

Question

In general, building on unpublished work of another author, which he may have told you in confidence, or you may encountered as a reviewer or by word of mouth or other informal means, is not an acceptable practice. Does the same apply if the author has uploaded his work on arXiv?

Here's a specific scenario: Suppose I submit a paper for review to a journal. The paper has results that could possibly be extended, however, the idea for extension is either not obvious or is not compatible with the theme of the paper, so I may plan to keep it for a later paper. If I put the paper on arXiv would it be considered ok if someone else extends my results? This would ruin the theme of the other paper I wanted to write with the extension.

If this happened to me it would make me uncomfortable, as if my confidence has been breached. I don't know if this feeling is justified. This feeling stems from my view of arXiv. My personal theory is that arXiv is an informal medium, unlike journals and conferences, and in that sense it is no different from other channels of informal communication. It allows for structured and faster dissemination of work (thus inform the ongoing work of others), in addition to establishing priority of results, but that does not grant it a formal status. I think an author should exercise restraint in extending results from arXiv, and apply the same standards that he would if he were to encounter the work through classical informal channels.

Of course, this is just my theory. I would like to hear what others have to say about this issue.

Answer 1

I understand your logic, but I do not think that many people treat papers on arXiv in that way. Papers on arXiv are public unlike the papers you reviewed or the work you just heard by personal communication. As such, extending the results in a paper on arXiv is a perfectly acceptable practice as long as the second paper cites the paper on arXiv properly.

Answer 2

I feel like publishing something on arXiv is an act of "publishing" nonetheless.

It has the advantages of publishing. These include: getting your work out there (with more visibility than if simply posting it on your webpage , which requires people be looking for your articles), getting people interested in it, and perhaps even getting citations before the paper has gone through the long and grueling process of being validated by a journal. It also had the advantage of showing that this is your result, thus preventing anybody from "preempting" you.

But then it also has the "disadvantage" that someone may run with your idea, and extend it, possibly even obtaining a result that is more significant than your own.

This does not mean publishing things on arXiv should be seen as a risk. It means that publishing on arXiv, like anywhere else, should be a well-thought decision. Note that with all the turmoil surrounding official journals, and common practices, it seems we are slowly evolving towards archives such as arXiv getting increasing legitimacy.

So to answer your question: an article on arXiv is not "unpublished" at all (unlike papers you get to read as a referee, or which a colleague has shared with you in confidence---those are other issues altogether), so yes, there is generally nothing wrong with extending results.

Answer 3

As the other replies say, putting your paper on a public archive is a way of making it public, and so other researchers may build on the results.

One thing to do when you want to make your work public, but you're also working on extensions that will take more time to establish, is to explicitly write in the paper you publish that you have an ongoing work on those extensions. This at least lets people know what the situation is. You can't control what people are going to do with this information, but I think that reasonable people would not immediately start working on the same extensions, and, if at some point they do want to work on these extensions, they are likely to contact you to check what's going on.

Of course, this "ongoing work" statement should be used only when it's appropriate and there's really an ongoing work that is likely to be completed within a reasonable amount of time.

Answer 4

In mathematics, it is very common to build on papers that are on arXiv. Indeed, this is more or less the whole point of arXiv: to disseminate work faster than the journal publication cycle, so that the community can immediately begin to make use of it, which includes extending it. If you wait until the paper is published, you will be well behind the curve.

I don't think your reservations about building on "unpublished" work are applicable here. At least in mathematics, preprints are generally posted on arXiv at the same time as they are submitted for publication; they are thus already in the "final" form intended by the authors. Of course it may be that further revisions take place before publication (perhaps as a result of the peer review process), but the arXiv version is usually perfectly suitable as a foundation for future work.

However, a few extra caveats are in order:

• Since a paper from arXiv has not passed peer review yet, you should probably be even more diligent than usual about checking its results before using them. If you find errors or unclear points, it would behoove you to alert the authors, since they may be able to be corrected before publication.

• Conversely, you should also keep an eye out for updates to the preprint which may contain important corrections.

• You may certainly cite the arXiv preprint in your own paper. However, you should check for a published version before final acceptance of your own (and again when correcting the proofs), and update the citation if necessary.

Answer 5

There is however one more important distinction between arXiv and journals/conferences. Whatever is published in journals/conferences is final and can no more be revised. It is as if the author has said all that he has to say on the topic and it is now fair game for others to say what they want.

Whether this is a meaningful distinction depends on the circumstances: it's common to write a series of papers on a given topic, in which case no single paper contains all the author has to say on the topic.

The other answers have clearly explained how the communities that use the arXiv view this matter. However, I'd go a step further: a system where people could stake out credit for unpublished work while preventing other people from building on it would be bad for science. Currently, this is the trade-off researchers face. You can keep your ideas private for as long as you want, to give you time to exploit them fully. However, you run the risk that someone else may independently discover them. In that case, the best you can hope for is to share the credit if you promptly make your work public (and make a convincing case that it was independent). On the other hand, if you circulate a paper publicly, then you no longer need to worry about credit, but anyone can use your ideas. This is very reasonable: credit is not awarded for being smart or accomplished, but rather for contributing to the community, and that contribution is the ideas. Being able to use and build on these ideas is why they are so valuable, and this is the justification for awarding credit.

The alternative to this trade-off would be to allow researchers to claim credit while temporarily reserving the exclusive right to do follow-up work. In that case, the optimal strategy would be clear: make a public announcement as soon as you cross the neecessary threshold, but then delay finalizing the work for as long as possible. This would be very disruptive for the research community.

In practice, things are slightly less black and white than I've described them above. It can be considered rude to compete too strenuously without a very good reason. If you come up with a minor idea building on someone else's work in progress, it's common to donate it to them rather than competing, and a major idea may lead to a collaboration. However, this is a matter of politeness rather than ethics.

Answer 6

There is a risk in releasing your results early (since others may build on your work before you do it), but there is also a risk in not doing it (since an article doing the exact same thing may appear before you make your preprint public).

There are also other advantages in publishing early (and, more in general, making your work known in conferences before it is fully peer-reviewed):

• instead of "stealing your ideas", people may contact you and propose a collaboration.
• if people know your work, they can cite you.

That is why I suggest releasing early, at least in my field (numerical linear algebra --- which, I realize, is a quite friendly and relaxed field, so your mileage may vary).

Answer 7

I upvoted all the earlier answers... and want to reiterate certain points. First, arXiv (and stable personally-maintained web-sites, e.g., at universities) should count as "publication" in the sense of making the information available, which is the ideal point of research.

Yes, there is some conflict with "the other" point, namely, staking a claim, getting credit, getting a job, getting tenure. But, as others have pointed out, it would have a bad effect if one could "stake a claim" without divulging details, thereby stifling others' work (if they needed to hope to get some "credit" also).

(There has always been an analogous hazard for beginners, all the worse pre-internet, that senior people can be "rumored" to be "working on something", circulated in mimeo to friends, etc. Mercifully, this is now diminished, I think, exactly because there's no excuse for limiting circulation.)

As Anon Math and others said, of course it is more civil to offer collaboration rather than to compete directly, etc., and this is vastly more consistent with our ideal of advancing knowledge (rather than mere self-advancement, as necessary as some of the latter may be).

In summary, then, yes, a relative novice risks having their ideas taken up by more experienced people. Yes, in particular, some people are able to rationalize that non-refereed papers need not be cited at all! This is, of course, barbaric. :)

Answer 8

A few points to add -

Why publish things on arxiv? So that others know your results. This is only relevant if you can get others to start using your results. It'll increase your citation count and speed up progress. You run the risk of getting people who publish things you want to do in the future.

What to do about the risk? You can explicitly state you plan on doing X. If I'm already doing X, that might light a fire and get me to quickly finish. Or I might try to collaborate with you. If I'm not already doing X, would it be wise for me to compete with you? I don't know how far along you are, but you're almost surely ahead of me - would I really put in the effort to try to catch up with such a high risk of being scooped? No.

So if you tell people you plan to do an extension the real risk you run is that someone else who is already doing it might speed up their work, not that someone will come along and start up quickly.